INDEX
Explanations
mentions of historical sites and their architectural features
New Auto-Interp
Negative Logits
æĬľ
-0.18
Doll
-0.15
ạn
-0.14
á»ķ
-0.14
velt
-0.14
ossil
-0.14
å®ħ
-0.14
wart
-0.14
Exhibit
-0.13
tu
-0.13
POSITIVE LOGITS
fres
0.31
cycle
0.26
scenes
0.25
dep
0.24
cycles
0.24
Scenes
0.24
alt
0.24
panel
0.24
panels
0.24
Cycle
0.23
Activations Density 0.061%